RESEARCH ARTICLE

A new level of plasticity: Drosophila smooth-like testes musclescompensate failure of myoblast fusionJessica Kuckwa, Katharina Fritzen, Detlev Buttgereit, Silke Rothenbusch-Fender and Renate Renkawitz-Pohl*

ABSTRACTThe testis of Drosophila resembles an individual testis tubule ofmammals. Both are surrounded by a sheath of smooth muscles,which in Drosophila are multinuclear and originate from a pool ofmyoblasts that are set aside in the embryo and accumulate on thegenital disc later in development. These muscle stem cells start todifferentiate early during metamorphosis and give rise to all musclesof the inner male reproductive system. Shortly before the genital discand the developing testes connect, multinuclear nascent myotubesappear on the anterior tips of the seminal vesicles. Here, we show thatadhesion molecules are distinctly localized on the seminal vesicles;founder cell (FC)-like myoblasts express Dumbfounded (Duf) andRoughest (Rst), and fusion-competent myoblast (FCM)-like cellsmainly express Sticks and stones (Sns). The smooth but multinuclearmyotubes of the testes arose by myoblast fusion. RNAi-mediatedattenuation of Sns or both Duf and Rst severely reduced the numberof nuclei in the testes muscles. Duf and Rst probably actindependently in this context. Despite reduced fusion in all of theseRNAi-treated animals, myotubes migrated onto the testes, testeswere shaped and coiled, muscle filaments were arranged as in thewild type and spermatogenesis proceeded normally. Hence, thetestes muscles compensate for fusion defects so that the myofibresencircling the adult testes are indistinguishable from those of the wildtype and male fertility is guaranteed.

KEY WORDS: Male fertility, Dumbfounded, Roughest, Sticks andstones, Hibris

INTRODUCTIONFunction and formation of multinuclear myofibres are largelyconserved within the animal kingdom. Numerous studies invertebrates and Drosophila melanogaster have revealed thatmultinuclear striated myotubes arise by myoblast fusion (Abmayrand Pavlath, 2012). Despite the similarities, only vertebrates possessmuscle stem cells, i.e. satellite cells, that allow growth andregeneration of muscles after injury (Wozniak et al., 2005). InDrosophila, the stem cells most similar to satellite cells are theadult muscle precursor cells (Figeac et al., 2007), which allowmodification of larval muscles into templates for dorsal longitudinalindirect flight muscles during metamorphosis (Roy andVijayRaghavan, 1998). These stem cells are set aside duringembryogenesis and amplify mitotically in the larvae (Bate et al.,

1991). Most of them are associated with imaginal discs asadepithelial cells in late third instar larvae. During metamorphosis,they build the adult musculature of the legs, thorax, and female andmale reproductive organs.

Muscle development in the Drosophila embryo has been studiedintensively. In the embryo, heterotypic myoblasts recognize andadhere to each other. After signal transduction from the cell surfaceinto the cell via adaptor proteins, F-actin reorganizes at the site ofcell-cell contact, the opposing membranes are vesiculated andcytoplasmic continuity is established (Haralalka and Abmayr, 2010;Önel et al., 2014). In this process, several molecular players relevantfor the formation of multinuclear myofibres have functionalredundancies (Bonn et al., 2013; Duan et al., 2012; Hakeda-Suzukiet al., 2002; Hornbruch-Freitag et al., 2011). Well-studied examplesof redundancy duringmyoblast fusion are cell adhesionmolecules ofthe immunoglobulin superfamily (IgSF), namely Dumbfounded[Duf; also known as Kin of irre (Kirre)], and Roughest (Rst), whichare expressed in founder cells (FCs) (Bate, 1990; Ruiz-Gómez et al.,2000). The genes encoding Duf and Rst are localized in the sameregion on the genome (St Pierre et al., 2014) and only the deletion ofboth leads to lack of fusion and embryonic lethality before sarcomereformation. Expression of Duf or Rst alone can rescue the deletionphenotype (Ruiz-Gómez et al., 2000; Strünkelnberg et al., 2001).Fusion-competentmyoblasts (FCMs) express Sticks and stones (Sns)and Hibris (Hbs). Loss of Sns leads to a nearly complete block offusion, whereas Hbs seems to be less essential (Bour et al., 2000;Dworak et al., 2001; Shelton et al., 2009).

All muscles of theDrosophilamale reproductive system originatefrom adepithelial cells of the sexually dimorphic genital disc(Ahmad and Baker, 2002; Estrada et al., 2003; Kozopas et al.,1998). During metamorphosis, parts of the genital disc differentiateinto the prospective seminal vesicle (vs) and the paragonia (pg)(Fig. 1A). The epithelial cells of the seminal vesicle and thedeveloping testes connect to each other so that muscle precursorscan migrate from the seminal vesicles onto the testes (Fig. 1B).Evidence from transplantation experiments and cultures of pupaltestes indicates that the connection between the seminal vesiclesand testes is essential for outgrowth and shaping of the testes(Gärtner et al., 2014; Kozopas et al., 1998; Nanda et al., 2009; Stern,1941a,b). Different types of muscles can be found around the innermale genitalia, specifically multinuclear smooth-like myofibressurrounding the testes, multinuclear striated muscles of the spermpump and a number of mononuclear striated muscles (Susic-Junget al., 2012). In contrast to striated muscles, smooth muscles lack theregular arrangement in a repetitive pattern of sarcomeres withZ-discs and regular pattern of Myosins in the middle and F-actinlinked to the Z-disc (Au, 2004). Smooth muscle cells are aheterogeneous group and they are well studied in mammals (Aliet al., 2005; Matsumoto and Nagayama, 2012). By contrast, allmuscles of the female reproductive organs are mononuclear andstriated (Hudson et al., 2008).Received 22 May 2015; Accepted 28 November 2015

Philipps-Universitat Marburg, Fachbereich Biologie, Entwicklungsbiologie,Karl-von-Frisch Strasse 8, Marburg 35043, Germany.

*Author for correspondence (renkawit@biologie.uni-marburg.de)

This is an Open Access article distributed under the terms of the Creative Commons AttributionLicense (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use,distribution and reproduction in any medium provided that the original work is properly attributed.

329

© 2016. Published by The Company of Biologists Ltd | Development (2016) 143, 329-338 doi:10.1242/dev.126730

DEVELO

PM

ENT

The molecular players responsible for the development of themuscles of the Drosophila male reproductive system and theirmechanisms of action remain mostly unstudied. Likewise, little isknown about the origin, development and function of mammalianperitubular myoid cells, i.e. the smooth muscle cells that enclose theseminiferous tubules of the testes where spermatozoa are generated(Svingen and Koopman, 2013). Recent analyses of the developmentof the different reproductive tract muscles of Drosophila haveprovided the first evidence that Duf, Sns and Hbs are relevant forarranging the multinuclear smooth-like muscles encircling the testes(Susic-Jung et al., 2012). In the present study, we focused on(1) how the multinuclear state of testes muscles is achieved; (ii) therole in this process of the IgSF proteins known from embryonicmyogenesis (in particular, we asked whether these adhesionmolecules act also redundantly during the development of thetestes muscles); and (3) how loss of the adhesion molecules affectsthe formation of the male reproductive system and its musculature.We show that the multinuclear smooth-like myofibres of the

testes arise by fusion of two cell types resembling FCMs and FCswith respect to their heterotypic expression of IgSF proteins. Allfour IgSF proteins were involved in this process, althoughapparently with a function different to that in embryonic myoblastfusion. Importantly, even when fusion was reduced, the malereproductive tract developed as in the wild type. In this case, thesmooth-like muscles displayed organized myofibres. Thus, ourresults reveal a high plasticity of smooth muscle formation.

RESULTSMultinuclear myoblasts are found on the developing seminalvesicles of male genital discsMyoblasts are characterized by the activity of theMef2 gene, whichencodes the highly conserved muscle-specific transcription factorMef2 (Bour et al., 1995; Lilly et al., 1994; Nguyen et al., 1994;Taylor et al., 1995). We established transgenic fly lines expressingUAS-mCD8-GFP under the control of Mef2-Gal4 allowing ectopic

expression in all myoblasts to follow myogenesis duringmetamorphosis with a live marker.

Myoblasts positive forMef2-driven mCD8-GFP expression werepresent on the genital disc already at the onset of metamorphosis(Fig. 2A). During the first hours of metamorphosis, these myoblastsproliferated and increased in number [Fig. 1B,C, and monitored byanti-phosphorylated histone H3 (Ser10) mitosis marker staining inFig. S1]. Myoblasts were found, 16 h after puparium formation(APF), in clusters on the prospective seminal vesicles, paragoniaand ejaculatory duct (Fig. 2C). By 24 h APF, the primordia of theparagonia and the seminal vesicles were clearly distinguishable, andmCD8-GFP-positive myoblasts were particularly prominent on theseminal vesicles (Fig. 2D). These myoblasts will mainly give rise tothe multinuclear smooth musculature of the testes and werestill mononuclear at 24 h APF (Fig. 2E-E″). We first observedbinuclear mCD8-GFP-positive myoblasts at 28 h APF, when theseminal vesicles grew closer towards the testes (Fig. 2F-F″). Here,mCD8-GFP expression is visible mainly in the cytoplasm andnot efficiently integrated into the plasma membrane. Thesemultinuclear myoblasts were restricted to the myoblast layer overthe most anterior part of the prospective seminal vesicle (Fig. 2F,arrowheads); the paragonium still contained only mononuclearmyoblasts. Up to this point, genital disc and testes are not attachedand Mef2-positive cells cannot be found on the developing testes(Bodenstein, 1950; Susic-Jung et al., 2012). AllMef2-positive cellsarriving on the testes at around 32 h APF are already multinuclear(Fig. 2G-G″). Those multinuclear mCD8-GFP-expressing cellscover the testes at 40 h APF (Fig. 2H-H″). Mononuclear Mef2-positive myoblasts were not observed on testes. These resultsshowed that myoblasts amplify during metamorphosis and firstbecome multinuclear on the prospective seminal vesicles of themale genital disc at around 28 h APF.We propose that further fusionleads to three, or rarely more, nuclei of the individual nascentmyoblasts. This multinuclearity appears to be completed when themyotubes reach the testes.

Fig. 1. Scheme of the origin and development of the testesmuscles. (A) Genital disc 20 h after puparium formation (APF) contains a pool of myoblasts on theprotruding seminal vesicles (vs). The paired testes (te) are free of myoblasts. (B) By 36 h APF, the testis and the developing seminal vesicle are fused.Multinucleated nascent myotubes migrate onto the testis. (C) The adult testis is surrounded by a sheath of multinuclear smooth-like muscles. Modified afterBodenstein (1950); Susic-Jung et al. (2012).

330

RESEARCH ARTICLE Development (2016) 143, 329-338 doi:10.1242/dev.126730

DEVELO

PM

ENT

The multinuclear state of testes muscles is probablyachieved by cell-cell fusionTo our knowledge, all multinuclear striated muscles in Drosophilaarise by fusion of myoblasts. However, the other mechanism ofgenerating multinuclear cells is by cytokinesis failure, i.e. nucleardivision without cytokinesis. This process occurs in the earlydevelopment of the Drosophila embryo and in mammalian heartmuscles (Foe et al., 1993; Li et al., 1996; Liu et al., 2010). Thus, weinvestigated whether syncytial smooth-like muscles of the testesdevelop by cell cycle arrest after nuclear division. We labelledreplicating DNA with the thymidine analogue 5-ethynyl-2′-deoxyuridine (EdU), which can be detected by covalent bindingof a fluorescent azide.Male pupae with Mef2-driven mCD8-GFP were injected with

EdU at 24 h APF, shortly before the first binuclear myoblastsformed. After incubation for 8 h, the pupae were dissected andfixed, EdU and GFP fluorescent labels were analysed. During the

8 h incubation, myoblasts became multinuclear and migrated fromthe male genital disc onto the testis. On the testis, myoblastsexpressing mCD8-GFP were detected in the basal area (Fig. 3A,left). At higher magnification, no EdU signal was detected in thenuclei of these nascent myotubes (Fig. 3A′). By contrast, germ cellsin the apical part of the testis (Fig. 3A, right) were clearly labelledwith EdU (Fig. 3A″, arrow). This demonstrated the progression ofdevelopment, as well as successful incorporation and detection ofEdU in dividing germ cells as an intrinsic positive control. On thebasis of these results, we concluded that the multinuclear state oftestes muscles arises in a cell-cycle-independent manner.

To further investigate how the multinuclear testes muscles areformed, we analysed the expression pattern of the rp298-lacZenhancer trap (Nose et al., 1998). This transgene follows the activityof the duf enhancer and is located to the nucleus by the NLS of thetransposase (Mlodzik andHiromi, 1992). In theDrosophila embryo,rp298-lacZ can be found in nuclei of FCs, but not in FCMs. After

Fig. 2. Myoblasts on the male genital disc become multinuclear right before they migrate onto the testes. (A) Myoblasts labelled by Mef2-drivenUAS-mCD8-GFP expression (green) on male genital discs at the onset of puparium formation, (B) 8 h APF, (C) 16 h APF and (D) 24 h APF. (E) AdditionalHoechst staining (blue) of DNA in themyoblast nuclei 24 h APF, (F) 28 h APF genital discs, and (G) 32 h and (H) 40 h testes. Single-channel magnifications of theboxed areas are displayed below. E,F,G and H, optical sections. Arrowheads, multinuclear myoblasts on the most anterior tips of the seminal vesicles or thetestes. pg, paragonium; vs, seminal vesicle; de, ejaculatory duct; te, testis. Scale bars: 20 µm.

Fig. 3. Testes myoblasts fuse to generatemultinuclear myotubes. (A-A″) Labelling ofreplicating DNA with 5-ethynyl-2′-deoxyuridine(EdU, red; arrow in A″) in pupal testis at 24 h APF,followed by incubation for 8 h. Myoblasts werelabelled by Mef2-driven UAS-mCD8-GFPexpression (green). Asterisk, hub region; magnifiedareas shown in A′ and A″ are indicated in A.Additional Hoechst staining (blue) of DNA (A′) in thebasal area and (A″) in the apical part of the testis.(B-B″) Visualization of rp298-lacZ (red) in nascentmyotubes (green) in pupal testes at 40 h APF; B is amerged photo of B′ and B″ with additional Hoechststaining. Arrowhead, rp298-lacZ positive nucleus.A′,A″ and B-B″, optical sections. Scale bars: 20 µm.

331

RESEARCH ARTICLE Development (2016) 143, 329-338 doi:10.1242/dev.126730

DEVELO

PM

ENT

fusion, all nuclei in the body wall muscles are positive for rp298-lacZ (Nose et al., 1998). By contrast, multinuclear myotubes oftestes at 40 h APF expressed rp298-lacZ predominantly in only onenucleus (Fig. 3B-B″, arrowhead), as has been described forbinuclear circular visceral muscles (Klapper et al., 2002). Thisfinding suggested that the nuclei of one testis muscle originate fromdifferent myoblasts.Taken together, these data indicated that nuclear division with

cytokinesis failure does not occur during the development ofDrosophila testes muscles. We thus concluded that a myoblastfusion process is obligatory in this context and thereforeinvestigated whether testes myoblast fusion is comparable to theFC- and FCM-based mechanism by which striated muscles areestablished in Drosophila.

Myoblasts on the prospective seminal vesicles showdifferences in IgSF expressionTo gain insights into the molecular players involved in testesmyoblast fusion, we checked for the presence of transcripts ofknown myogenesis-relevant genes in the myoblasts of male genitaldiscs at different time points during metamorphosis. We establisheda protocol to purify myoblasts from male genital discs. Total RNAof age-specific mCD8-GFP-positive myoblasts was isolated(Fig. 4A) and used for RT-PCR analyses (Fig. 4B, Table S1).Hoechst staining of DNA revealed the high purity of isolatedmyoblasts (Fig. 4A).Initially, we investigated whether distinct myoblasts similar to

embryonic FCMs and FCs exist on the genital disc. In the embryo,Notch/Delta-mediated lateral inhibition leads to specification ofFCs and FCMs (Baker and Schubiger, 1996; Carmena et al., 1995;Corbin et al., 1991). We confirmed active transcription of the geneencoding the Notch receptor and the gene encoding its ligandDelta in purified myoblasts (Fig. 4B), and lack of transcription ofthe spermatid specifically expressed negative control Transitionprotein like94D (Tpl94D) (Rathke et al., 2007). We also demonstratedtranscription of the FCM-specific determination factor lameduck (lmd) (Duan et al., 2001) and FC-specific transcription

factors (Fig. 4B, Table S1). Several, but not all, genes encodingtranscriptional regulators conferring FC or muscle identity in theembryo were transcribed in these myoblasts (Table S1). Thus, wehypothesize that there are FC- and FCM-like myoblasts on malegenital discs.

We further investigated whether myoblasts express the genesencoding IgSF proteins that initiate embryonic myoblast fusion.Using gene-specific oligonucleotides, we were able to amplifythe adhesion molecule transcripts duf, rst, sns and hbs frommyoblast RNA of genital discs 8, 16, 24 and 30 h APF (Fig. 4B),i.e. not only during fusion events, but also earlier. However, thismethod cannot distinguish between transcripts from testesmyoblasts and transcripts from myoblasts that give rise to othermuscles of the reproductive tract, such as multinuclear striatedmuscles of the sperm pump. Therefore, to determine the preciseexpression pattern and the subcellular localization of theseIgSF proteins, we used immunofluorescence staining. In theprospective seminal vesicles of genital discs at 24 h APF, anti-Duf (Fig. 4C), anti-Rst (Fig. 4D) and anti-Sns (Fig. 4E) signalswere detected in the membranes of Mef2-positive myoblasts(arrows). The adhesion molecules were distributed evenly overthe plasma membrane and not localized to specific regions; asmCD8-GFP expression was localized further into the myoblastthan IgSFs, we assume that mCD8 is mainly cytoplasmic in thesecells (see also Fig. 2). Only a subset of myoblasts was stainedwith each antibody; Duf and Rst appeared to be enriched inmyoblasts adjacent to the epithelium of the primordial seminalvesicles (Fig. 4C,D), whereas Sns accumulated in myoblastslying on the periphery of the tissue (Fig. 4E). Furthermore, wedid not observe colocalization of Sns and Duf (Fig. 4F,F′) or ofSns and Rst (Fig. 4G,G′) in double staining, which indicatedexpression of Sns in myoblasts other than those that expressedDuf and Rst. We hypothesize that these different subsets ofmyoblasts fuse to each other to generate smooth-like testesmuscles. Consequently, we refer to them as FC-like myoblastsand FCM-like cells. FC-like myoblasts express Duf and Rst,whereas FCM-like cells mainly express Sns.

Fig. 4. Adhesion molecules are transcribed andexpressed in myoblasts on the male genitaldisc. (A) Myoblasts expressing Mef2-drivenUAS-mCD8-GFP (green, arrowhead) purified usingmagnetic beads (double arrowhead). Hoechststaining of DNA (red). (B) RT-PCR results usingRNA from purified myoblasts at 8 to 24 h APF.(C-E) Staining in male genital discs at 24 h APF in aMef2-driven UAS-mCD8-GFP background withantibodies against (C) Duf, (D) Rst and (E) Sns.Arrows, myoblasts located on the prospectiveseminal vesicles. (F-G″) Double staining in malegenital discs at 24 h APF with antibodies against(F,F′) Sns and Duf and (G,G′) Sns and Rst. F′ andG′ are enlargements of boxed areas shown in F andG, respectively. F″ andG″ are schematic depictionsof the antibody distribution seen in F′ and G′. C-G′,optical sections. vs, seminal vesicle; pg,paragonium. Scale bars: 20 µm.

332

RESEARCH ARTICLE Development (2016) 143, 329-338 doi:10.1242/dev.126730

DEVELO

PM

ENT

Knockdown of Duf, Rst or Sns leads to a lower nuclei numberin testes musclesAs the adhesion molecules Duf, Rst and Sns were expressed, weanalysed whether their presence is essential for the formation ofmultinuclear testes muscles. As hbs transcript could be detected inmyoblasts isolated from the genital disc, we included hbs in thisfunctional study. We down-regulated duf, rst, sns and hbsspecifically in myoblasts using Mef2-Gal4-driven RNAinterference (RNAi). The adult-muscle-specific driver line 1151-Gal4 did not reveal any activity in the myoblasts of the male genitaldisc and was therefore excluded from further studies (data notshown). Because Duf and Rst can act in redundancy to each other,we additionally downregulated both duf and rst in a duf-RNAi;rst-RNAi double knockdown. For comparison, we expressed Dicer-2(Dcr-2) in the RNAi background, because it enhances the effect ofRNAi (Dietzl et al., 2007).As the muscle sheath tightly encircles the adult testis, single

muscles and their cell nuclei are difficult to visualize; we saw only1-2 nuclei on one side of the testes (Susic-Jung et al., 2012; Fig. 1C).First nascent myotubes reach the testes at 32 h APF (Fig. 2G) andare still individually distinguishable at 48 h APF; they did notsurround the testes at this time (Fig. 2H, 40 h APF). At these timepoints, only multinucleated nascent myotubes but no mononuclearmyoblasts were observed on the testes (Fig. 2E,F). We counted thenumber of nuclei in nascent myotubes on wild-type testes at 36 h,42 h and 48 h APF (Table S2). The number of nuclei did notincrease during ongoing metamorphosis, making further fusion onthe testes unlikely. We conclude that fusion mainly takes place onthe genital disc. Therefore, we analysed testes from pupae around42 h APF, because migrating nascent myotubes have not stretchedto their full size, which allows individual myotubes to be clearlyrecognized and the number of nuclei per muscle to be easilydetermined (Fig. 5A,B).In the wild type, the majority of nascent myotubes comprised 3

nuclei, and some had 2 or 4 nuclei (Fig. 5C); the average was 3.1

nuclei. In rare cases, mononuclear myotubes or myotubes with 5 or6 nuclei were observed. In myotubes in which duf, rst, hbs or snswere individually downregulated, the number of nuclei was reduced(Fig. 5C, see also Figs S2, S3 and Table S2 for controls and details).Co-expression of Dcr-2 with the RNAi constructs led to a moresevere reduction in nuclei number (Fig. 5C); expression of eitherduf-RNAi or rst-RNAi with Dcr-2 increased the number of mono-and binuclear myotubes from the observed 21% in the wild type to85% and 70%, respectively. The average number of nuclei declinedto 2 and 2.3, respectively. Knockdown of hbs resulted in an averagenumber of 2.4 nuclei, which was slightly lower than that of thecontrols. When both duf and rst were downregulated, the number ofnuclei was reduced further than in the single knockdowns. Again,co-expression of Dcr-2 led to a more severe reduction in the nucleinumber; mono- and binuclear muscles were observed in 93% of themyotubes, with an average nuclei count of about 1.6. Expression ofa sns-RNAi construct byMef2-Gal4 led to the most severe reductionin nuclei, with an average of about 1.1; more than 90% of myotubesremained mononuclear. Furthermore, these pupae did not developinto adults. Co-expression of Dcr-2 with sns-RNAi resulted in earlypupal lethality and therefore could not be analysed. From thesefindings, we concluded that the number of nuclei in testis musclesdepends on the IgSF proteins that initiate myoblast fusion in theembryo.

Development of the male reproductive tract andspermatogenesis appears normal despite a severe reductionin fusion efficiencyWe then analysed whether the nuclei number of testes musclesaffects functionality. We tested whether males in which duf, rst orhbs were downregulated were fertile. Three virgin wild-typefemales were mated with single males for 7 days; we tested up to30 RNAi knockdown males. Offspring were evaluated 14 days aftermating started. sns-RNAi males could not be analysed becausedepletion of sns led to pupal lethality.

Fig. 5. Knockdown of adhesion molecules leads to reduced nuclei number in testes muscles. (A,B) Visualization of muscles and nuclei using anti-Tropomyosin (green) and Hoechst staining (white) of (A) wild-type myotubes and (B) myotubes expressing Mef2-driven Dcr-2, duf-RNAi; rst-RNAi. Images aremerged photos of single optical sections. The outlined muscle is displayed in split channels in A′,B′. Arrowheads, muscle nuclei. Scale bars: 20 µm.(C) Determination of nuclei number of nascent myotubes in which duf, rst, duf;rst, hbs and snswere downregulated byMef2-driven RNAi and in wild-type testes atabout 42 h APF. In each case, 200 muscles (100% of the cells) were counted; the length of each colour-coded bar in each stack indicates the percentage ofmuscles with the specified number of nuclei (see Table S2 for details).

333

RESEARCH ARTICLE Development (2016) 143, 329-338 doi:10.1242/dev.126730

DEVELO

PM

ENT

Males with downregulated duf, rst or hbs, alone or in combinationwith Dcr-2 as well as the duf;rst double knockdown producedoffspring, which indicated normal fertility. Only males with Mef2-driven duf;rst double knockdown in the Dcr-2 background hadreduced fertility (Fig. 6A). Only 31% of single crossings producedoffspring, and the number of offspring was lower than in the othermales (Table S2 and Fig. S4); 69% were infertile. The reproductivetract of these infertile males had no visible blockages, constrictionsor other defects (Fig. 6B). The seminal vesicles were filled withmotile sperm (Fig. 6C, Movie 1); no obvious defects inspermatogenesis were detected. To identify the cause of reducedfertility in these flies, we first checked the mated wild-type femalesthat failed to produce progeny. In wild-type female reproductiveorgans, sperm are stored in the spermatheca (spt) and the tubularreceptacles (tr; Fig. 5D). The spermatheca and tubular receptacles offemales mated with RNAi knockdown males were free of sperm(compare Fig. 6D and E). Subsequently, we examined the fertility ofMef2-driven Dcr-2;duf-RNAi;rst-RNAi females by crossing themwith wild-type males. Their fertility was also severely reduced,showing that the effect was not sex-specific as would be expectedfor defects due to muscles of the testes. Furthermore, these malesand females were able to crawl and jump but were unable to fly(Table S2). As sperm production in the males is not disturbed andfemales display reduced fertility, we assume that the cause ofreduced fertility in these flies is a general problem in behaviour and/or movement rather than defective testes muscle development. Wealso conclude that except for Dcr-2;duf-RNAi;rst-RNAi, a reducednumber of nuclei in the testes muscles does not necessarily interferewith male fertility.

Knockdown of duf, rst or hbs leads to a wild-type adultmuscle pattern with a normal filament arrangementAs fertility was not affected by RNAi-mediated knockdown ofduf, rst or hbs, the question arose whether testes muscles areformed correctly despite a reduced nuclei number. We analysedthe overall shape and filament arrangement of adult reproductivetracts of RNAi knockdown males using F-actin and DNA staining.The testes of all RNAi knockdown flies had the characteristiccoiled morphology of the wild type, and as already mentioned, theadult reproductive tracts did not display any visible defects,constrictions or other alterations compared with the wild type(Fig. 6B, Dcr-2;duf-RNAi;rst-RNAi; other knockdown flies notshown). Furthermore, the muscles from hbs-RNAi, duf-RNAi, rst-RNAi and duf-;rst-RNAi knockdown males did not display obviousdefects. Fibre distribution and filament arrangements of the testesmuscle sheaths and the sperm pump musculature were normalwhen observed under light microscopy (Fig. 7, Dcr-2;duf-RNAi;rst-RNAi; other knockdown flies not shown) independent of theco-expression of Dcr-2. In addition, no defects were found in themuscle sheaths of paragonia, seminal vesicles or ejaculatory ductsof any of the tested RNAi knockdown flies (Fig. S5). Adult maleswith a reduced Sns level could not be analysed because sns-RNAiis pupal lethal.

Our results show that knockdown of IgSF proteins resulted in areduced nuclei number in testes muscles. Surprisingly, this did notaffect (1) the migration of nascent myotubes from the seminalvesicles onto the testes, (2) the morphogenesis of the testes from thelarval to the adult shape, (3) filament arrangement and integrity ofthe reproductive tract musculature and (4) the fertility of the males.

Fig. 6. RNAi-mediated knockdown of duf, rst or hbs in the reproductive tract musculature does not influence male fertility. (A) Fertility of duf, rst, duf;rstand hbs knockdown males compared with wild-type males. Error bars represent s.d. (B) Reproductive tracts of infertile males (merged photograph). (C) Seminalvesicles of infertile males are filled with mature sperm. (D) Wild-type females mated with wild-type males contain sperm in their tubular receptacles (arrows) orspermatheca. (E) Reproductive tracts of females of infertile crossings with RNAi knockdown males are free of sperm (arrows). spt, spermatheca; tr, tubularreceptacle; te, testis; vs, seminal vesicle; pg, paragonium; de, ejaculatory duct; sp, sperm pump. Scale bars: 10 µm in B, 20 µm in C-E.

334

RESEARCH ARTICLE Development (2016) 143, 329-338 doi:10.1242/dev.126730

DEVELO

PM

ENT

Hence, we propose that testes muscles possess mechanisms tocompensate reduced fusion.

DISCUSSIONDespite the differences between the diverse types of striatedmuscles of Drosophila, they share common mechanisms duringdevelopment. For example, the expression pattern of Duf suggeststhat Duf plays a similar role in the development of larval muscles(Ruiz-Gómez et al., 2000), leg muscles (Soler et al., 2004), flightmuscles (Gildor et al., 2012) and abdominal muscles (Dutta et al.,2004). The cellular adhesion mediated by the IgSF proteins andtheir signalling in myoblasts is essential for myoblast fusion andformation of striated muscles.

Smooth-like testes muscles arise from heterotypic fusion ofFC-like and FCM-like cells, localized in two distinct myoblastlayersIt is unclear how the special smooth-like testes muscles obtain theirmultinuclear state. In general, cells can fuse or skip cleavage afternuclear division to become multinuclear (Gentric and Desdouets,2014; Lacroix and Maddox, 2012). It was shown that garland cellnephrocytes can overcome fusion defects by cell division withoutcytokinesis (Zhuang et al., 2009). In the male reproductive organs,the binuclear epithelial cells of the paragonia arise by cytokinesisskipping (Taniguchi et al., 2014). By contrast, our studydemonstrated that multinuclearity in testes muscles is achieved bymyoblast fusion. Only one of the nuclei of these small syncytia ispositive for rp298-lacZ, as we observed previously for the bi-nucleated circular visceral muscles in the embryo (Klapper et al.,2002). We proposed that rp298-lacZ – reflecting Duf activity – is nolonger transcribed after fusion and thus, no new β-galactosidase issynthesized in the cytoplasm and the protein cannot therefore beimported into the other nuclei of these small syncytia. Thus,downregulation of duf transcription could limit the degree of fusion,resulting in small syncytia. By contrast, in the somatic mesoderm,Menon et al. (2005) proposed that reshuffling Duf and its adaptor

protein Rols to the membrane provides a mechanism that regulatesthe rate of fusion to yield larger syncytia in agreement with theobservation that all nuclei of the syncytia are positive for rp298(Menon et al., 2005).

Furthermore, we found that characteristic IgSF molecules ofembryonic myoblast fusion were expressed distinctly in thesemyoblasts. This is in agreement with previous expression patterns ofsns and duf reporter constructs (Susic-Jung et al., 2012). The FCM-and FC-like cell status of two different populations of myoblasts issupported by our finding of only one rp298-positive nucleus in thesmall syncytia, suggesting that this nucleus of one testis muscle celloriginates from a different myoblast to the others.

Antibody staining revealed that FC-like myoblasts express Dufand Rst and form a basal layer directly adjacent to the epithelium,whereas the FCM-like cells express Sns and lie more in theperiphery (schematized in Fig. 8). In the embryo, adhesionmolecules are mainly visible at the opposing membranes ofgrowing myotubes and FCMs (Kesper et al., 2007; Rochlin et al.,2010; Sens et al., 2010; Önel et al., 2011). By contrast, testesmyoblasts expressed adhesion molecules along their entire surface,which indicates discernible differences in the formation of thedifferent muscle types.

Duf and Rst might act independently, not redundantly, tocreate smooth-like myofibresKnockdown of duf, rst or sns during metamorphosis led to reducedfusion processes in each muscle. Thus, we conclude that myoblastfusion during development of multinuclear smooth-like testesmuscles shares common molecular players with previously studiedmyoblast fusion processes of Drosophila and vertebrates. However,we observed a distinct mode of action of Duf and Rst during testes-relevant myoblast fusion. Fusion was also reduced when only duf orrst was knocked down. In the embryo, only a double knockoutleads to defects, whereas the reintroduction of either Duf or Rst issufficient to rescue the double-mutant phenotype (Ruiz-Gómezet al., 2000; Strünkelnberg et al., 2001). In thoracic muscledevelopment during metamorphosis, duf-knockdown myoblastsfuse like the wild type, but duf-RNAi;rst-RNAi double-knockdownflight muscles have severe fusion defects, which also indicates aredundant function of Duf and Rst (Gildor et al., 2012). By contrast,Duf and Rst cannot completely substitute for each other in thearrangement of ommatidia during eye formation (Ramos et al.,1993; Reiter et al., 1996). Our data demonstrate that in testesmyoblast fusion, the phenotype of the single RNAi knockdown isenhanced when both genes are knocked down simultaneously,which indicates an additive effect.

Fusion is restricted to themyoblasts at the anterior tip of theseminal vesiclesWe observed that testes myoblast fusion was restricted to a smallarea of the most anterior tip of the prospective seminal vesicles(Figs 2 and 8). Furthermore, our RNAi data indicated that properexpression of Duf, Rst and Sns is crucial for testes myoblastfusion to proceed. As adhesion molecules were expressed inneighbouring FC- and FCM-like cells long before fusion occurs,we conclude that these molecules alone are not sufficient to triggerfusion of adjacent myoblasts. In other fusion processes, keyregulators called fusogens are required and are sufficient for fusionto proceed (Aguilar et al., 2013). To date, the fusogen inmyoblasts of Drosophila has not been identified. Our resultsdemand an accurate spatially and temporally regulated expressionof a potential fusogen within the testes myoblasts. Therefore, there

Fig. 7. Knockdown of duf and rst during development does not lead todefects in filament organization in adult reproductive muscles. Filamentarrangement of (A,C) wild-type and (B,D) duf-RNAi;rst-RNAi doubleknockdown in (A,B) smooth-like testes muscles and sarcomere pattern of(C,D) multinuclear sperm pump muscles. F-Actin (red) was visualized withPhalloidin; Hoechst dye (white in A and B; blue in C and D) was used to labelDNA in nuclei. In C and D, anti-Mef2 antibody (green) was used to detectmuscle nuclei. All photographs are optical sections. Scale bars: 20 µm.

335

RESEARCH ARTICLE Development (2016) 143, 329-338 doi:10.1242/dev.126730

DEVELO

PM

ENT

might be an extrinsic signal that triggers fusion solely in themyoblasts over the seminal vesicles. Alternatively, fusion might berepressed in all other myoblasts. Clarification of these possibilitieswill be a focus of future research.

Compensation of fusion failure leads to correct filamentorganization of testes muscles, shaping of the testes andmale fertilityIn Drosophila and vertebrates, reduced fusion of striated musclescan lead to severe defects or embryonic lethality (Abmayr andPavlath, 2012; Rochlin et al., 2010; Önel and Renkawitz-Pohl,2009). Furthermore, defective morphogenesis of the reproductivesystem can induce male sterility (Linnemannstons et al., 2014).However, despite a reduction in fusion in the knockdown testesmuscles, no obvious defects in formation of the reproductive tractand progression of spermatogenesis were detected in the resultingadult males. The seminal vesicles were always connected to thetestes, and the myotubes migrated onto the testes. Even the filamentarrangement in these flies appeared normal in light microscopy,independent of the level of the fusion block (Fig. 7, Fig. 8B). Thus,we assume that testes muscles remain functional even if fusion isreduced. We conclude that progression of myoblast fusion is notrequired for further development of the testes muscles, includingattachment, filament production and arrangement. Kreisköther et al.(2006) showed that sarcomeres of striated larval muscles areestablished very late in embryogenesis after muscles attached to theepidermis. This process seems to proceed independent of myoblastfusion in some muscles also in fusion mutants (Drysdale et al.,1993). Similar observations have been reported for striated musclesin zebrafish; mutants of the IgSF genes jamb and jamc displayfusion defects during fast-twitch muscle development, whereaselongation and sarcomere formation are not disturbed (Powell andWright, 2011).

In summary, our results indicate that testes muscles compensatefor fusion defects in a manner not previously reported. We suggestthat testes muscles have an enormous potential for increasing theplasticity of smooth muscles. We propose that this plasticitycompensates for the lack of satellite cells in Drosophila and theirability to regenerate muscle defects and thereby safeguardsreproductive capability.

MATERIALS AND METHODSFly stocksFlies were maintained and RNAi was performed in Drosophila standardmedium at 25°C. w1118 (BL6326) was used as the wild-type control. Thefollowing transgenic flies were used: rp298-lacZ (Nose et al., 1998),Mef2-Gal4 (Ranganayakulu et al., 1995), UAS-Dcr-2;Mef2-Gal4 (BL25756),UAS-mCD8-GFP (BL32186), UAS-duf-RNAi (V3111), UAS-rst-RNAi(V951), UAS-hbs-RNAi (V40898) and UAS-sns-RNAi (V108577). It hasbeen shown previously that these RNAi constructs mediate efficientknockdown with other driver lines (Gildor et al., 2012; Machado et al.,2011; Susic-Jung et al., 2012). BL flies were obtained from BloomingtonDrosophila Stock Center; V flies were ordered from the Vienna DrosophilaRNAi Center.

Tissue preparation and immunofluorescenceTo acquire pupae of a defined age, white prepupae were collected at 0 hAPF and aged on a moistened filter. Pupae or adult flies were dissectedin a drop of phosphate-buffered saline (PBS; 0.13 mM NaCl, 7 mMNa2HPO4, 3 mM NaH2PO4) under a stereomicroscope. For antibodystaining, tissues were fixed in 3.7% formaldehyde in PBS for 20 min,primary antibody was incubated overnight at 4°C and secondaryantibody was added to the specimens at room temperature for 1 h.Hoechst 33258 (3 µg/ml; Sigma-Aldrich, 94403) was used to label DNAand Atto565-phalloidin (4 nmol/l; Sigma-Aldrich, 94072) was used tostain F-actin.

The following antibodies were used: anti-GFP (1:2000; Abcam, ab6556),anti-tropomyosin (1:1000; Abcam, ab50567), anti-phosphorylated histone

Fig. 8. Scheme of the distinct layers ofFC-like and FCM-like cells that fuse onthe seminal vesicle to generate thetestes muscles. (A) Myoblasts on theseminal vesicle (vs) are separated into twolayers. FC-like myoblasts (blue) expressDuf and Rst and lie adjacent to theepithelium of the seminal vesicle, whereasFCM-like cells (yellow) express Sns andappear to be located on the periphery. Inheterotypic fusion events, FC- and FCM-like cells fuse to form multinuclear nascentmyotubes. These myotubes then migrateonto the testes (te). The adult testis istightly encircled by multinuclear muscles.(B) The knockdown of IgSF moleculesimpairs the fusion process. Nascentmyotubes with a reduced nuclei numberstill migrate from the seminal vesicle ontothe testes and are able to enclose the adulttestes as in the wild type.

336

RESEARCH ARTICLE Development (2016) 143, 329-338 doi:10.1242/dev.126730

DEVELO

PM

ENT

H3 (Ser10) mitosis marker (1:500; Millipore, 06-570), anti-Mef2 (1:500;kindly provided by Hanh T. Nguyen, University of Erlangen, Germany),anti-Duf and anti-Rst (1:500 and 1:50, respectively; both gifts from Karl-Friedrich Fischbach, University of Freiburg, Germany), anti-β-Gal (1:1000;Biotrend, RGA1-45A-Z) and anti-Sns (1:250; generated by PinedaAntibody Service, Berlin). For fluorescent immunohistochemistry, thefollowing secondary antibodies were used: anti-rat Alexa Fluor 488(Jackson ImmunoResearch Laboratories), anti-rabbit DyLight488 (VectorLaboratories) and anti-guinea pig Cy2 (Jackson ImmunoResearchLaboratories).

EdU labelling of pupal tissueTo detect proliferation, Click-iT EdU Alexa Fluor 594 (Molecular Probes,Invitrogen) was used. EdU (0.1 µl of 0.5 mM) with Toluidine Blue (0.25%),for a better visualization in PBS, was injected into the dorso-medial part ofthe pupal thorax as described elsewhere (Ito and Hotta, 1992) at 24 h APFand incubated for 8 h at 25°C. After dissection and fixation of the testes,anti-GFP staining was performed. EdU was detected following thesupplier’s instructions.

RNA isolation from myoblastsCells were isolated following the protocol of Wang et al. (2006) withmodifications. Briefly, 20 UAS-mCD8-GFP;Mef2-Gal4 genital discs of 8,16, 24 and 30 h APF were dissected in cold dissociation buffer (Sigma,C1544) andkept cold until further processing.Washed genital discswere thendissociated by elastase treatment (5 µg/ml; Sigma, E0127). Myoblasts werepurified using magnetic beads [Dynabeads Mouse CD8 (Lyt2), Invitrogen]according to the supplier’s instructions. Subsequently, total RNAwas isolatedfrom these myoblasts using the RNAqueous-Micro total RNA Isolation Kit(Ambion) following instructions therein. RT-PCR was performed using theOneStep RT-PCR Kit (Qiagen) and oligonucleotides listed in Table S3.

Image acquisition and processingConventional fluorescent images and optical sections were gathered with aZeiss AxioObserver Z.1 inverse microscope with an attached ApoTome thatwas used for structured illumination microscopy. Images were merged usingAdobe Photoshop CS 5.1, plates were arranged in GIMP 2.8 and charts weregenerated in Microsoft Excel 2010.

AcknowledgementsWe thank Susanne Onel and Anja Rudolf for fruitful discussions and critical readingof the manuscript, Katja Gessner for organizational assistance and graphic designexpertise, Peer Fender for help with graphics, Ljubinka Cigoja for injectingDrosophila pupae, Hanh Nguyen and Karl-Friedrich Fischbach for antibodies, XinChen for methodical support and Karen A. Brune for linguistic revision. We receivedfly strains from Vienna Drosophila RNAi Center, Austria. Stocks obtained from theBloomington Drosophila Stock Center (NIH P40OD018537) were used in this study.

Competing interestsThe authors declare no competing or financial interests.

Author contributionsJ.K., D.B. and R.R.-P. conceived and designed the experiments. J.K., K.F., D.B. andS.R.-F. performed the experiments, and J.K. and R.R.-P. wrote the manuscript.

FundingThis work was supported by the Deutsche Forschungsgemeinschaft [Re 628/16-1and GRK 1216]. Deposited in PMC for immediate release.

Supplementary informationSupplementary information available online athttp://dev.biologists.org/lookup/suppl/doi:10.1242/dev.126730/-/DC1

ReferencesAbmayr, S. M. and Pavlath, G. K. (2012). Myoblast fusion: lessons from flies andmice. Development 139, 641-656.

Aguilar, P. S., Baylies, M. K., Fleissner, A., Helming, L., Inoue, N., Podbilewicz,B., Wang, H. andWong, M. (2013). Genetic basis of cell-cell fusion mechanisms.Trends Genet. 29, 427-437.

Ahmad, S. M. andBaker, B. S. (2002). Sex-specific deployment of FGF signaling inDrosophila recruits mesodermal cells into the male genital imaginal disc.Cell 109,651-661.

Ali, F., Pare, P. D. and Seow, C. Y. (2005). Models of contractile units and theirassembly in smooth muscle. Can. J. Physiol. Pharmacol. 83, 825-831.

Au, Y. (2004). The muscle ultrastructure: a structural perspective of the sarcomere.Cell. Mol. Life Sci. 61, 3016-3033.

Baker, R. and Schubiger, G. (1996). Autonomous and nonautonomous Notchfunctions for embryonic muscle and epidermis development in Drosophila.Development 122, 617-626.

Bate, M. (1990). The embryonic development of larval muscles in Drosophila.Development 110, 791-804.

Bate, M., Rushton, E. and Currie, D. A. (1991). Cells with persistent twistexpression are the embryonic precursors of adult muscles in Drosophila.Development 113, 79-89.

Bodenstein, D. (1950). The postembryonic development of Drosophila. In Biologyof Drosophila (ed. M. Demerec), pp. 275-367. New York: John Wiley & Sons, Inc.

Bonn, B. R., Rudolf, A., Hornbruch-Freitag, C., Daum, G., Kuckwa, J., Kastl, L.,Buttgereit, D. and Renkawitz-Pohl, R. (2013). Myosin heavy chain-like localizesat cell contact sites during Drosophila myoblast fusion and interacts in vitro withRolling pebbles 7. Exp. Cell Res. 319, 402-416.

Bour, B. A., O’Brien, M. A., Lockwood, W. L., Goldstein, E. S., Bodmer, R.,Taghert, P. H., Abmayr, S. M. and Nguyen, H. T. (1995). Drosophila MEF2, atranscription factor that is essential for myogenesis. Genes Dev. 9, 730-741.

Bour, B., Chakravarti, M., West, J. and Abmayr, S. (2000). Drosophila SNS, amember of the immunoglobulin superfamily that is essential for myoblast fusion.Genes Dev. 14, 1498-1511.

Carmena, A., Bate, M. and Jimenez, F. (1995). Lethal of scute, a proneural gene,participates in the specification of muscle progenitors during Drosophilaembryogenesis. Genes Dev. 9, 2373-2383.

Corbin, V., Michelson, A. M., Abmayr, S. M., Neel, V., Alcamo, E., Maniatis, T.and Young, M. W. (1991). A role for the Drosophila neurogenic genes inmesoderm differentiation. Cell 67, 311-323.

Dietzl, G., Chen, D., Schnorrer, F., Su, K.-C., Barinova, Y., Fellner, M., Gasser,B., Kinsey, K., Oppel, S., Scheiblauer, S. et al. (2007). A genome-widetransgenic RNAi library for conditional gene inactivation in Drosophila. Nature448, 151-156.

Drysdale, R., Rushton, E. and Bate, M. (1993). Genes required for embryonicmuscle development in Drosophila melanogaster A survey of the X chromosome.Dev. Biol. 202, 276-295.

Duan, H., Skeath, J. B. and Nguyen, H. T. (2001). Drosophila Lame duck, a novelmember of the Gli superfamily, acts as a key regulator of myogenesis bycontrolling fusion-competent myoblast development. Development 128,4489-4500.

Duan, R., Jin, P., Luo, F., Zhang, G., Anderson, N. andChen, E. H. (2012). Group IPAKs function downstream of Rac to promote podosome invasion duringmyoblast fusion in vivo. J. Cell Biol. 199, 169-185.

Dutta, D., Anant, S., Ruiz-Gomez, M., Bate, M. and VijayRaghavan, K. (2004).Founder myoblasts and fibre number during adult myogenesis in Drosophila.Development 131, 3761-3772.

Dworak, H. A., Charles, M. A., Pellerano, L. B. and Sink, H. (2001).Characterization of Drosophila hibris, a gene related to human nephrin.Development 128, 4265-4276.

Estrada, B., Casares, F. and Sanchez-Herrero, E. (2003). Development of thegenitalia in Drosophila melanogaster. Differentiation 71, 299-310.

Figeac, N., Daczewska, M., Marcelle, C. and Jagla, K. (2007). Muscle stem cellsand model systems for their investigation. Dev. Dyn. 236, 3332-3342.

Foe, V. E., Odell, G. M. and Edgar, B. A. (1993). Mitosis and morphogenesis in theDrosophila embryo: point and counterpoint. In The Development of Drosophilamelangaster (ed. M. Bate and A. Martinez-Arias), pp. 149-300. Cold SpringHarbor; New York: Cold Spring Harbor Laboratory Press.

Gartner, S. M., Rathke, C., Renkawitz-Pohl, R. and Awe, S. (2014). Ex vivo cultureof Drosophila pupal testis and single male germ-line cysts: dissection, imaging,and pharmacological treatment. J. Vis. Exp. 91, 51868.

Gentric, G. and Desdouets, C. (2014). Polyploidization in liver tissue.Am. J. Pathol. 184, 322-331.

Gildor, B., Schejter, E. D. and Shilo, B.-Z. (2012). Bidirectional Notch activationrepresses fusion competence in swarming adult Drosophila myoblasts.Development 139, 4040-4050.

Hakeda-Suzuki, S., Ng, J., Tzu, J., Dietzl, G., Sun, Y., Harms, M., Nardine, T.,Luo, L. and Dickson, B. J. (2002). Rac function and regulation during Drosophiladevelopment. Nature 416, 438-442.

Haralalka, S. and Abmayr, S. M. (2010). Myoblast fusion in Drosophila. Exp. CellRes. 316, 3007-3013.

Hornbruch-Freitag, C., Griemert, B., Buttgereit, D. and Renkawitz-Pohl, R.(2011). Drosophila Swiprosin-1/EFHD2 accumulates at the prefusion complexstage during Drosophila myoblast fusion. J. Cell Sci. 124, 3266-3278.

Hudson, A. M., Petrella, L. N., Tanaka, A. J. and Cooley, L. (2008). Mononuclearmuscle cells in Drosophila ovaries revealed by GFP protein traps. Dev. Biol. 314,329-340.

Ito, K. andHotta, Y. (1992). Proliferation pattern of postembryonic neuroblasts in thebrain of Drosophila melanogaster. Dev. Biol. 149, 134-148.

337

RESEARCH ARTICLE Development (2016) 143, 329-338 doi:10.1242/dev.126730

DEVELO

PM

ENT

Kesper, D. A., Stute, C., Buttgereit, D., Kreiskother, N., Vishnu, S., Fischbach,K.-F. and Renkawitz-Pohl, R. (2007). Myoblast fusion in Drosophilamelanogaster is mediated through a fusion-restricted myogenic-adhesivestructure (FuRMAS). Dev. Dyn. 236, 404-415.

Klapper, R., Stute, C., Schomaker, O., Strasser, T., Janning, W., Renkawitz-Pohl, R. and Holz, A. (2002). The formation of syncytia within the visceralmusculature of the Drosophila midgut is dependent on duf, sns and mbc. Mech.Dev. 110, 85-96.

Kozopas, K. M., Samos, C. H. and Nusse, R. (1998). DWnt-2, a Drosophila Wntgene required for the development of the male reproductive tract, specifies asexually dimorphic cell fate. Genes Dev. 12, 1155-1165.

Kreiskother, N., Reichert, N., Buttgereit, D., Hertenstein, A., Fischbach, K.-F.and Renkawitz-Pohl, R. (2006). Drosophila rolling pebbles colocalises andputatively interacts with alpha-Actinin and the Sls isoform Zormin in the Z-discs ofthe sarcomere and with Dumbfounded/Kirre, alpha-Actinin and Zormin in theterminal Z-discs. J. Muscle Res. Cell Motil. 27, 93-106.

Lacroix, B. and Maddox, A. S. (2012). Cytokinesis, ploidy and aneuploidy.J. Pathol. 226, 338-351.

Li, F., Wang, X., Capasso, J. M. and Gerdes, A. M. (1996). Rapid transition ofcardiac myocytes from hyperplasia to hypertrophy during postnatal development.J. Mol. Cell. Cardiol. 28, 1737-1746.

Lilly, B., Galewsky, S., Firulli, A. B., Schulz, R. A. and Olson, E. N. (1994).D-MEF2: a MADS box transcription factor expressed in differentiating mesodermandmuscle cell lineages during Drosophila embryogenesis.Proc. Natl. Acad. Sci.USA 91, 5662-5666.

Linnemannstons, K., Ripp, C., Honemann-Capito, M., Brechtel-Curth, K.,Hedderich, M. and Wodarz, A. (2014). The PTK7-related transmembraneproteins off-track and off-track 2 are co-receptors for Drosophila Wnt2 required formale fertility. PLoS Genet. 10, e1004443.

Liu, Z., Yue, S., Chen, X., Kubin, T. and Braun, T. (2010). Regulation ofcardiomyocyte polyploidy and multinucleation by CyclinG1. Circ. Res. 106,1498-1506.

Machado, M. C. R., Octacilio-Silva, S., Costa, M. S. A. and Ramos, R. G. P.(2011). rst transcriptional activity influences kirre mRNA concentration in theDrosophila pupal retina during the final steps of ommatidial patterning. PLoS ONE6, e22536.

Matsumoto, T. and Nagayama, K. (2012). Tensile properties of vascular smoothmuscle cells: bridging vascular and cellular biomechanics. J. Biomech. 45,745-755.

Menon, S. D., Osman, Z., Chenchill, K. and Chia, W. (2005). A positive feedbackloop between Dumbfounded and Rolling pebbles leads to myotube enlargementin Drosophila. J. Cell Biol. 169, 909-920.

Mlodzik, M. and Hiromi, Y. (1992). Enhancer trap method in Drosophila: itsapplication to neurobiology. In Methods in Neurosciences: Gene Expression inNeural Tissues (ed. P. M. Conn) pp. 397-414. Amsterdam: Elsevier.

Nanda, S., DeFalco, T. J., Loh, S. Y. H., Phochanukul, N., Camara, N., VanDoren, M. and Russell, S. (2009). Sox100B, a Drosophila group E Sox-domaingene, is required for somatic testis differentiation. Sex Dev. 3, 26-37.

Nguyen, H. T., Bodmer, R., Abmayr, S. M., McDermott, J. C. and Spoerel, N. A.(1994). D-mef2: aDrosophilamesoderm-specific MADS box-containing genewitha biphasic expression profile during embryogenesis. Proc. Natl. Acad. Sci. USA91, 7520-7524.

Nose, A., Isshiki, T. and Takeichi, M. (1998). Regional specification of muscleprogenitors in Drosophila: the role of the msh homeobox gene.Development 125,215-223.

Onel, S.-F. and Renkawitz-Pohl, R. (2009). FuRMAS: triggering myoblast fusion inDrosophila. Dev. Dyn. 238, 1513-1525.

Onel, S.-F., Dottermusch, C., Sickmann, A., Buttgereit, D. and Renkawitz-Pohl,R. (2011). Role of the actin cytoskeleton within FuRMAS during Drosophilamyoblast fusion and first functionally conserved factors in vertebrates. In CellFusions: Regulation and Control (ed. L.-I. Larsson) pp. 139-170. Berlin: Springer.

Onel, S.-F., Rust, M. B., Jacob, R. and Renkawitz-Pohl, R. (2014). Tetheringmembrane fusion: common and different players in myoblasts and at the synapse.J. Neurogenet. 28, 302-315.

Powell, G. T. and Wright, G. J. (2011). Jamb and jamc are essential for vertebratemyocyte fusion. PLoS Biol. 9, e1001216.

Ramos, R. G., Igloi, G. L., Lichte, B., Baumann, U., Maier, D., Schneider, T.,Brandstatter, J. H., Frohlich, A. and Fischbach, K. F. (1993). The irregular

chiasm C-roughest locus of Drosophila, which affects axonal projections andprogrammed cell death, encodes a novel immunoglobulin-like protein. GenesDev. 7, 2533-2547.

Ranganayakulu, G., Zhao, B., Dokidis, A., Molkentin, J. D., Olson, E. N. andSchulz, R. A. (1995). A series of mutations in the D-MEF2 transcription factorreveal multiple functions in larval and adult myogenesis in Drosophila. Dev. Biol.171, 169-181.

Rathke, C., Baarends, W. M., Jayaramaiah-Raja, S., Bartkuhn, M., Renkawitz, R.and Renkawitz-Pohl, R. (2007). Transition from a nucleosome-based to aprotamine-based chromatin configuration during spermiogenesis in Drosophila.J. Cell Sci. 120, 1689-1700.

Reiter, C., Schimansky, T., Nie, Z. and Fischbach, K. F. (1996). Reorganization ofmembrane contacts prior to apoptosis in the Drosophila retina: the role of theIrreC-rst protein. Development 122, 1931-1940.

Rochlin, K., Yu, S., Roy, S. and Baylies, M. K. (2010). Myoblast fusion: when ittakes more to make one. Dev. Biol. 341, 66-83.

Roy, S. and VijayRaghavan, K. (1998). Patterning muscles using organizers: larvalmuscle templates and adult myoblasts actively interact to pattern the dorsallongitudinal flight muscles of Drosophila. J. Cell Biol. 141, 1135-1145.

Ruiz-Gomez, M., Coutts, N., Price, A., Taylor, M. V. and Bate, M. (2000).Drosophila dumbfounded: a myoblast attractant essential for fusion. Cell 102,189-198.

Sens, K. L., Zhang, S., Jin, P., Duan, R., Zhang, G., Luo, F., Parachini, L. andChen, E. H. (2010). An invasive podosome-like structure promotes fusion poreformation during myoblast fusion. J. Cell Biol. 191, 1013-1027.

Shelton, C., Kocherlakota, K. S., Zhuang, S. and Abmayr, S. M. (2009). Theimmunoglobulin superfamily member Hbs functions redundantly with Sns ininteractions between founder and fusion-competent myoblasts. Development136, 1159-1168.

Soler, C., Daczewska, M., Da Ponte, J. P., Dastugue, B. and Jagla, K. (2004).Coordinated development of muscles and tendons of the Drosophila leg.Development 131, 6041-6051.

St Pierre, S. E., Ponting, L., Stefancsik, R. and McQuilton, P. (2014). FlyBase102-advanced approaches to interrogating FlyBase. Nucleic Acids Res. 42,D780-D788.

Stern, C. (1941a). The growth of testes in Drosophila. I. The relation between vasdeferens and testis within various species. J. Exp. Zool. 87, 113-158.

Stern, C. (1941b). The growth of the testes in Drosophila. II. The nature ofinterspecific differences. J. Exp. Zool. 87, 159-180.

Strunkelnberg, M., Bonengel, B., Moda, L., Hertenstein, A., de Couet, H.,Ramos, R. and Fischbach, K. (2001). rst and its paralogue kirre act redundantlyduring embryonic muscle development in Drosophila. Development 128,4229-4239.

Susic-Jung, L., Hornbruch-Freitag, C., Kuckwa, J., Rexer, K.-H., Lammel, U.and Renkawitz-Pohl, R. (2012). Multinucleated smooth muscles andmononucleated as well as multinucleated striated muscles develop duringestablishment of the male reproductive organs of Drosophila melanogaster.Dev. Biol. 370, 86-97.

Svingen, T. and Koopman, P. (2013). Building the mammalian testis: origins,differentiation, and assembly of the component cell populations. Genes Dev. 27,2409-2426.

Taniguchi, K., Kokuryo, A., Imano, T., Minami, R., Nakagoshi, H. and Adachi-Yamada, T. (2014). Isoform-specific functions ofMud/NuMAmediate binucleationof Drosophila male accessory gland cells. BMC Dev. Biol. 14, 46.

Taylor, M. V., Beatty, K. E., Hunter, H. K. and Baylies, M. K. (1995). DrosophilaMEF2 is regulated by twist and is expressed in both the primordia anddifferentiated cells of the embryonic somatic, visceral and heart musculature.Mech. Dev. 50, 29-41.

Wang, X., Bo, J., Bridges, T., Dugan, K. D., Pan, T.-C., Chodosh, L. A. andMontell, D. J. (2006). Analysis of cell migration using whole-genome expressionprofiling of migratory cells in the Drosophila ovary. Dev. Cell 10, 483-495.

Wozniak, A. C., Kong, J., Bock, E., Pilipowicz, O. and Anderson, J. E. (2005).Signaling satellite-cell activation in skeletal muscle: markers, models, stretch, andpotential alternate pathways. Muscle Nerve 31, 283-300.

Zhuang, S., Shao, H., Guo, F., Trimble, R., Pearce, E. and Abmayr, S. M. (2009).Sns and Kirre, the Drosophila orthologs of Nephrin and Neph1, direct adhesion,fusion and formation of a slit diaphragm-like structure in insect nephrocytes.Development 136, 2335-2344.

338

RESEARCH ARTICLE Development (2016) 143, 329-338 doi:10.1242/dev.126730

DEVELO

PM

ENT